Abstract
Background:
Food insecurity may affect both access to food and the quantities acquired by households. However, evidence on its relationship with the acquisition of specific food groups in Peru remains limited.
Objective:
To evaluate the association between food insecurity severity and the probability and quantity of food acquisition among Peruvian households.
Methods:
A cross-sectional study was conducted using data from 32,053 households participating in the 2025 National Household Survey (ENAHO). Food insecurity was assessed using the Food Insecurity Experience Scale (FIES; 0–8 points). Per capita acquisition of fruits, vegetables, legumes, cereals, pseudocereals and cereal-based foods, fish and seafood, processed meats, and sugar-sweetened beverages was estimated. Modified Poisson regression models were used to estimate adjusted prevalence ratios (aPRs), and Gamma regression models with a log link were used to estimate adjusted mean ratios (aMRs), accounting for the complex survey design.
Results:
Each 1-point increase in FIES was associated with 2.6% lower fruit acquisition (aMR = 0.974; 95% CI: 0.968–0.980) and 1.3% lower vegetable acquisition (aMR = 0.987; 95% CI: 0.982–0.992). Higher FIES severity was also associated with lower probabilities of acquiring fish and seafood, processed meats, and carbonated soft drinks. Cereals, pseudocereals, and cereal-based foods showed only minimal positive associations in continuous models, while categorical analyses showed no monotonic gradient across severity bands.
Conclusion:
Greater food insecurity severity was associated with selective contraction of household food acquisition, particularly for fruits and vegetables, rather than a uniform shift across all food groups.
1 Introduction
Food insecurity reflects the constraints faced by individuals and households in regularly accessing sufficient, safe, and nutritious food (Carvajal-Aldaz et al., 2022). It can also be understood as a continuum of severity that begins with worry or uncertainty about food access and may progress to compromises in dietary quality and variety, reductions in the amount of food consumed, and, in its most severe forms, food deprivation (Hanson and Connor, 2014; Hernández-Vásquez et al., 2026). The Food Insecurity Experience Scale (FIES) was specifically developed to capture this progression through experiences related to constrained food access (FAO, 2016). In response to these constraints, households may adopt different coping strategies, such as reducing the quantity or variety of foods, prioritizing lower-cost products, or modifying their usual food acquisition decisions (Brunet et al., 2026; Gupta and Narine, 2025). These responses are nutritionally relevant because food insecurity has consistently been associated with poorer overall diet quality, particularly among adults (Hanson and Connor, 2014).
However, much of the available evidence has focused on the relationship between food insecurity and dietary intake, diet quality, or dietary patterns (Hutchinson and Tarasuk, 2022; Ihab et al., 2012; Kent et al., 2024), whereas household-level food acquisition has received comparatively less attention. Food can enter the household not only through market purchases but also through own production, self-supply, in-kind payments, gifts or transfers, and formal food-assistance programs. Household consumption and expenditure surveys can capture several of these monetary and nonmonetary pathways and thereby provide an approximation of food availability within the household (FAO–World Bank, 2018). In a low-income district of Lima, families experiencing food insecurity reported using both formal food-assistance programs and informal food resources, including buying food on credit and receiving food from outside the household, with use of these resources tending to increase with food-insecurity severity (Brewer et al., 2021). Previous studies have also linked food insecurity to changes in food acquisition practices and to lower consumption or availability of fruits, vegetables, fresh or minimally processed foods, and animal-source foods (Betancourt-Núñez et al., 2023; Coletro et al., 2023; Ilori et al., 2024).
These changes may be shaped by economic constraints, as some nutritionally valuable foods, such as fruits and vegetables, have relatively high costs per unit of energy compared with cereal- or sugar-based products (Drewnowski, 2010). However, household responses may not be limited to substituting nutritious foods with lower-cost alternatives; some food groups may decline, others may remain relatively stable, and some may decrease as part of a broader contraction of the household food basket. In this study, food basket contraction refers to reductions in the probability of acquiring specific food groups and/or in the quantities acquired, rather than to a direct measure of dietary diversity or overall diet quality. In this context, assessing only whether a household acquires a given food may be insufficient, because that food may remain present in the basket while being acquired in progressively smaller quantities. Distinguishing between the probability of acquisition and the quantity acquired therefore provides a more precise characterization of how increasing food insecurity severity is reflected in household food acquisition decisions.
In Peru, this issue is particularly relevant given persistent socioeconomic disparities in diet. National studies have documented low fruit and vegetable consumption and marked inequalities by socioeconomic status; between 2014 and 2019, adequate consumption of these foods was consistently concentrated among wealthier groups (Hernández-Vásquez et al., 2022). More recent national data also indicate that average fruit and vegetable intake among Peruvian adults remains well below international recommendations (Guzman-Vilca et al., 2025). In addition, the National Household Survey (ENAHO) has been used to examine food expenditure and the nutritional quality of foods available in Peruvian households (Bonilla-Aguilar and Bernabe-Ortiz, 2023). More recently, the inclusion of the FIES in ENAHO 2025 enabled evaluation of its metric properties and estimation of the magnitude of food insecurity in the country (Hernández-Vásquez et al., 2026). However, it remains insufficiently understood how increasing food insecurity severity translates into specific changes in the acquisition of individual food groups among Peruvian households, particularly whether these changes occur through a lower probability of acquisition, reduced quantities acquired, or both.
Therefore, this study aimed to evaluate the association between food insecurity severity and both the probability and quantity of acquisition of specific food groups among Peruvian households using nationally representative data from ENAHO 2025. Fruits, vegetables, and legumes were examined as the primary food groups. Cereals, pseudocereals, and cereal-based foods, fish and seafood, processed meats, and carbonated soft drinks were additionally assessed. We hypothesized that greater food-insecurity severity would be associated with lower food acquisition, particularly of fruits and vegetables, and that jointly examining acquisition probability and quantity would identify food-group-specific patterns consistent with selective contraction of the household food basket.
2 Materials and methods
2.1 Study design and data source
A secondary cross-sectional analysis was conducted using data from the 2025 ENAHO, administered by Peru’s National Institute of Statistics and Informatics (INEI). ENAHO uses a probabilistic, area-based, stratified, and multistage sampling design, covering both urban and rural areas and allowing nationally representative estimates for the Peruvian population (INEI, 2025b).
Publicly available microdata were obtained from the Summary Database and modules 130 (Food Insecurity), 200 (Household Member Characteristics), 300 (Education), 601 (Food and Beverage Expenditures), and 700 (Social Programs), available through the INEI microdata repository (INEI, 2025a). Databases were linked using the common identifiers CONGLOME, VIVIENDA, and HOGAR. Individual-level information from modules 200 and 300 was first processed and aggregated at the household level before being merged with the food acquisition data.
2.2 Study population
The Summary Database initially included 33,702 households. Households with incomplete information on any of the eight items of the Food Insecurity Experience Scale (FIES) were excluded, as were 12 households for which the educational level of the household head could not be classified. The final analytical sample comprised 32,053 households. The household was the unit of analysis.
2.3 Food insecurity
Food insecurity was assessed using the eight FIES items included in ENAHO 2025. Affirmative responses were coded as 1 and negative responses as 0, and the items were summed to obtain a score ranging from 0 to 8, with higher scores indicating greater food insecurity severity. According to FAO methodology, the raw FIES score represents an ordinal measure of severity (FAO, 2016). In the main models, the FIES score was analyzed as a continuous ordinal exposure. For descriptive and sensitivity analyses, scores were additionally grouped into 0, 1–2, 3–5, and 6–8 to represent increasing levels of severity; these groupings were not intended to correspond to official FAO food insecurity categories. The metric properties of the FIES in the same ENAHO 2025 dataset were recently evaluated using the Rasch model (Hernández-Vásquez et al., 2026).
2.4 Household food acquisition
Household food acquisition was derived from Module 601, which records foods and beverages obtained for household consumption through purchases and other sources, including own production, self-supply, in-kind payments, gifts, and social programs. For each food item, total acquisition was calculated as the sum of the imputed and annualized quantities purchased (I601B2) and obtained from other sources (I601D2), both expressed in kilograms. Quantities were aggregated by food group and divided by household size to obtain kg/person/year. Food-group total records ending in 00 were excluded to avoid double counting.
Foods were classified using P601A, P601X, and PRODUC61 according to a prespecified classification dictionary. Item-level records, including eligible open product records, were retained within the corresponding food groups. Primary outcomes were fruits, vegetables, and legumes. Cereals, pseudocereals, and cereal-based foods were assessed as an additional food group, while fish and seafood, processed meats, and carbonated soft drinks were also examined. Prepared dishes were not disaggregated because standardized recipes were unavailable. The complete classification is provided in Supplementary Table S1.
For each food group, two complementary outcomes were analyzed: (1) any acquisition, defined as a total acquired quantity > 0, and (2) the quantity acquired among households with positive acquisition, expressed as kg/person/year. This approach allowed us to distinguish between the probability that a household acquired a given food group and the quantity obtained among households that did so.
2.5 Covariates
Covariates were selected a priori based on their conceptual relevance to both food insecurity and household food acquisition. Sex, age, and educational attainment of the household head were included to capture sociodemographic characteristics related to household resources and food decision-making. Household size and the presence of children aged <5 years or adults aged ≥65 years were included because household composition influences food needs and allocation. Poverty status represented direct economic constraints, whereas urban/rural residence and geographic domain captured differences in food environments, market access, and territorial context. Month of interview was included to account for temporal and seasonal variation in food availability and acquisition.
Poverty status was defined according to the official classification provided by INEI. Participation in food-assistance programs was obtained from Module 700. Because program participation is targeted toward households with specific vulnerabilities and may alter how food insecurity translates into household acquisition, it was evaluated as a potential effect modifier rather than interpreted as an independent causal exposure.
2.6 Statistical analysis
Analyses were conducted in R version 4.4.1 (R Foundation for Statistical Computing, Vienna, Austria). All analyses accounted for the complex sampling design of ENAHO by incorporating the household expansion factor (FACTOR07), strata, and clusters, allowing population-level estimates and standard errors that reflect the survey design (Lumley, 2004). Household characteristics were summarized using unweighted frequencies and weighted percentages with 95% confidence intervals (95% CIs). Because food acquisition quantities were right-skewed, they were summarized using weighted medians and interquartile ranges.
The association between FIES severity and food acquisition was evaluated using a two-part modeling approach. In the first part, the probability of acquiring each food group was estimated using Poisson regression with a log link and robust variance, with results expressed as adjusted prevalence ratios (aPRs) and 95% CIs, following the modified Poisson approach. In the second part, among households with positive acquisition, acquired quantities were modeled using generalized linear models with a Gamma distribution and log link, with results expressed as adjusted mean ratios (aMRs) and 95% CIs.
Primary models included the FIES score ranging from 0 to 8 and were adjusted for the prespecified covariates. For the quantity outcomes presented in Table 1, effect modification by poverty status, urban/rural residence, and participation in food-assistance programs was evaluated using FIES-by-modifier interaction terms, with stratum-specific aMRs and interaction p values reported. These effect-modification analyses were considered exploratory. FIES was also modeled categorically using raw-score bands (0, 1–2, 3–5, and 6–8) to examine departures from a monotonic pattern. As a sensitivity analysis, the influence of extremely high acquisition values was examined for these quantity outcomes. The 99.9th percentile was calculated from the distribution of per capita quantities among households with positive acquisition, excluding zero values from cutoff estimation; quantity models were then repeated after excluding positive observations above the food-group-specific threshold. Extreme observations were retained in the primary analyses. All tests were two-sided, and p < 0.05 was considered statistically significant.
Table 1
| Analysis | Fruits, aMR (95% CI) | Vegetables, aMR (95% CI) | Legumes, aMR (95% CI) | Processed meats, aMR (95% CI) | Carbonated soft drinks, aMR (95% CI) | Cereals, pseudocereals, and cereal-based foods, aMR (95% CI) |
|---|---|---|---|---|---|---|
| Panel A. Effect modification | ||||||
| Poverty status | ||||||
| Non-poor | 0.969 (0.962–0.975) | 0.989 (0.983–0.994) | 0.998 (0.990–1.005) | 0.988 (0.975–1.001) | 0.990 (0.978–1.002) | 1.008 (1.004–1.012) |
| Poor/extreme poverty | 0.989 (0.978–1.000) | 0.976 (0.966–0.987) | 0.992 (0.977–1.008) | 0.980 (0.955–1.007) | 0.980 (0.958–1.003) | 0.996 (0.988–1.004) |
| P for interaction | 0.001 | 0.038 | 0.523 | 0.631 | 0.451 | 0.010 |
| Area of residence | ||||||
| Urban | 0.971 (0.964–0.978) | 0.990 (0.984–0.995) | 1.001 (0.993–1.009) | 0.986 (0.974–0.999) | 0.987 (0.975–0.999) | 1.007 (1.003–1.011) |
| Rural | 0.990 (0.979–1.001) | 0.972 (0.964–0.979) | 0.973 (0.962–0.985) | 0.986 (0.962–1.012) | 0.994 (0.977–1.011) | 0.993 (0.986–1.000) |
| P for interaction | 0.004 | <0.001 | <0.001 | 0.995 | 0.547 | <0.001 |
| Food-assistance participation | ||||||
| No | 0.969 (0.962–0.977) | 0.991 (0.985–0.996) | 1.001 (0.992–1.010) | 0.983 (0.966–1.000) | 0.983 (0.970–0.996) | 1.011 (1.006–1.016) |
| Yes | 0.985 (0.977–0.993) | 0.978 (0.971–0.986) | 0.989 (0.978–0.999) | 0.985 (0.967–1.003) | 0.991 (0.971–1.011) | 0.997 (0.991–1.002) |
| P for interaction | 0.004 | 0.006 | 0.073 | 0.873 | 0.480 | <0.001 |
| Panel B. FIES modeled categorically | ||||||
| FIES score 0 | 1 (Reference) | 1 (Reference) | 1 (Reference) | 1 (Reference) | 1 (Reference) | 1 (Reference) |
| FIES score 1–2 | 0.902 (0.874–0.932) | 0.984 (0.956–1.012) | 0.983 (0.945–1.023) | 0.936 (0.865–1.010) | 0.943 (0.893–0.997) | 1.044 (1.021–1.068) |
| FIES score 3–5 | 0.882 (0.854–0.911) | 0.987 (0.962–1.013) | 0.987 (0.947–1.029) | 0.949 (0.887–1.020) | 0.884 (0.836–0.934) | 1.076 (1.054–1.099) |
| FIES score 6–8 | 0.827 (0.790–0.865) | 0.885 (0.850–0.922) | 0.971 (0.923–1.023) | 0.881 (0.804–0.966) | 0.954 (0.874–1.040) | 1.003 (0.974–1.033) |
| Panel C. Excluding values >99.9th percentile | ||||||
| Per 1-point increase in FIES | 0.974 (0.968–0.980) | 0.987 (0.982–0.992) | 0.996 (0.990–1.003) | 0.986 (0.974–0.998) | 0.988 (0.978–0.999) | 1.005 (1.001–1.009) |
Effect modification and sensitivity analyses of food insecurity and household food acquisition in ENAHO 2025.
Values are adjusted mean ratios (aMRs) with 95% confidence intervals for food quantities among households with acquisition >0. Panel A presents stratum-specific estimates and P values for interaction. Panel B models FIES using raw-score bands, with score 0 as the reference. Panel C excludes observations above the food-group-specific 99.9th percentile. Models were adjusted for the prespecified covariates and accounted for the ENAHO complex survey design. Effect-modification analyses should be interpreted as exploratory. aMR, adjusted mean ratio; CI, confidence interval; FIES, Food Insecurity Experience Scale; ENAHO, Encuesta Nacional de Hogares (Peruvian National Household Survey).
2.7 Ethical considerations
This study involved a secondary analysis of publicly available ENAHO microdata without personal identifiers accessible to the researchers; therefore, additional approval from an ethics committee was not required. ENAHO is responsible for the relevant ethical procedures, including obtaining informed consent from participants.
3 Results
Table 2 presents the weighted characteristics of the 32,053 households included in the analytic sample. Most households were headed by men (61.5%) and by individuals aged 30–59 years (58.6%), while secondary education was the most common educational level among household heads (41.7%). Nearly three-quarters of households had two to five members (74.8%); 50.4% included at least one individual younger than 18 years, and 32.4% included an adult aged ≥65 years. Overall, 20.4% of households were living in poverty, including 3.6% in extreme poverty, and 30.4% participated in at least one food-assistance program. Most households were located in urban areas (82.0%), with 53.0% residing on the Coast, 34.7% in the Highlands, and 12.3% in the Jungle.
Table 2
| Characteristic | Unweighted n | Weighted % (95% CI) |
|---|---|---|
| Sex | ||
| Male | 20,253 | 61.5 (60.8–62.3) |
| Female | 11,800 | 38.5 (37.7–39.2) |
| Age, years | ||
| 15–29 | 1,615 | 5.1 (4.7–5.4) |
| 30–59 | 18,594 | 58.6 (57.8–59.5) |
| 60–74 | 8,245 | 25.8 (25.0–26.5) |
| ≥75 | 3,599 | 10.5 (10.0–11.0) |
| Educational attainment | ||
| No formal education/preschool/basic special | 1,883 | 5.5 (5.1–5.8) |
| Primary education | 9,935 | 28.4 (27.7–29.1) |
| Secondary education | 12,310 | 41.7 (40.8–42.6) |
| Higher education | 7,925 | 24.3 (23.5–25.2) |
| Household size | ||
| 1 member | 5,588 | 15.8 (15.2–16.3) |
| 2–5 members | 23,534 | 74.8 (74.2–75.5) |
| ≥6 members | 2,931 | 9.4 (8.9–9.9) |
| Child aged <5 years in household | ||
| No | 26,727 | 84.1 (83.5–84.7) |
| Yes | 5,326 | 15.9 (15.3–16.5) |
| Child/adolescent aged <18 years in household | ||
| No | 16,012 | 49.6 (48.8–50.5) |
| Yes | 16,041 | 50.4 (49.5–51.2) |
| Adult aged ≥65 years in household | ||
| No | 21,637 | 67.6 (66.8–68.4) |
| Yes | 10,416 | 32.4 (31.6–33.2) |
| Poverty status | ||
| Extreme poverty | 1,138 | 3.6 (3.3–3.9) |
| Poverty | 4,714 | 16.8 (16.1–17.5) |
| Non-poor | 26,201 | 79.6 (78.8–80.4) |
| Participation in food-assistance programs | ||
| No | 21,221 | 69.6 (68.9–70.4) |
| Yes | 10,832 | 30.4 (29.6–31.1) |
| Area of residence | ||
| Urban | 22,721 | 82.0 (81.5–82.6) |
| Rural | 9,332 | 18.0 (17.4–18.5) |
| Natural region | ||
| Coast | 13,435 | 53.0 (51.5–54.4) |
| Highlands | 11,609 | 34.7 (33.2–36.2) |
| Jungle | 7,009 | 12.3 (11.5–13.2) |
Weighted characteristics of households included in the analytic sample, Peru, ENAHO 2025.
Values are presented as unweighted frequencies and survey-weighted percentages with 95% confidence intervals. Estimates account for the complex sampling structure of ENAHO 2025 using household sampling weights, strata, and primary sampling units. Percentages may not sum to exactly 100% because of rounding. ENAHO, Encuesta Nacional de Hogares (Peruvian National Household Survey).
Table 3 shows that increasing FIES severity was accompanied by lower acquisition of several food groups. Median fruit acquisition decreased progressively from 82.5 kg/person/year among households with a FIES score of 0 to 57.5 kg/person/year among those scoring 6–8, while vegetable acquisition declined from 66.5 to 51.0 kg/person/year. Legume acquisition remained comparatively stable across FIES raw-score bands. Higher FIES scores were also accompanied by lower proportions of households acquiring fish and seafood, processed meats, and carbonated soft drinks. In contrast, acquisition of cereals, pseudocereals, and cereal-based foods remained high across FIES categories, with no clear monotonic decline in the quantities acquired.
Table 3
| Food group and indicator | Overall n = 32,053 | FIES 0 n = 13,738 | FIES 1–2 n = 6,088 | FIES 3–5 n = 8,733 | FIES 6–8 n = 3,494 |
|---|---|---|---|---|---|
| Fruits | |||||
| Households acquiring, weighted % | 98.1 | 97.9 | 98.4 | 98.4 | 97.4 |
| Amount acquired, median (IQR), kg/person/year | 70.6 (39.2–120.5) | 82.5 (47.0–138.6) | 68.1 (38.8–111.9) | 62.0 (34.9–105.0) | 57.5 (30.0–100.6) |
| Vegetables | |||||
| Households acquiring, weighted % | 95.2 | 93.6 | 96.1 | 97.1 | 94.5 |
| Amount acquired, median (IQR), kg/person/year | 63.3 (37.5–99.8) | 66.5 (39.9–103.0) | 64.6 (38.1–101.1) | 63.0 (38.4–100.1) | 51.0 (28.5–83.6) |
| Legumes | |||||
| Households acquiring, weighted % | 74.8 | 73.3 | 77.2 | 77.2 | 69.9 |
| Amount acquired, median (IQR), kg/person/year | 11.0 (6.0–18.0) | 11.4 (6.0–18.0) | 10.8 (6.0–17.6) | 11.8 (6.0–18.6) | 10.1 (6.0–17.6) |
| Cereals, pseudocereals, and cereal-based foods | |||||
| Households acquiring, weighted % | 98.5 | 98.0 | 98.8 | 99.1 | 98.4 |
| Amount acquired, median (IQR), kg/person/year | 92.2 (59.4–130.2) | 89.5 (56.7–127.7) | 93.3 (61.1–130.9) | 96.8 (63.9–136.1) | 86.6 (56.3–124.2) |
| Fish and seafood | |||||
| Households acquiring, weighted % | 81.4 | 84.2 | 81.9 | 78.1 | 78.3 |
| Amount acquired, median (IQR), kg/person/year | 12.6 (6.1–24.0) | 14.0 (7.1–25.2) | 12.5 (6.0–22.8) | 12.0 (5.6–21.4) | 12.6 (5.8–24.3) |
| Processed meats | |||||
| Households acquiring, weighted % | 32.8 | 37.6 | 32.7 | 28.3 | 26.9 |
| Amount acquired, median (IQR), kg/person/year | 2.5 (1.2–5.0) | 2.7 (1.4–5.0) | 2.4 (1.2–4.5) | 2.4 (1.2–4.8) | 2.1 (1.0–4.2) |
| Carbonated soft drinks | |||||
| Households acquiring, weighted % | 45.4 | 48.0 | 47.7 | 41.9 | 40.9 |
| Amount acquired, median (IQR), kg/person/year | 12.6 (6.3–25.2) | 15.1 (8.0–25.2) | 12.6 (6.0–25.2) | 12.0 (6.0–24.0) | 12.0 (6.0–25.2) |
Household food acquisition patterns according to food insecurity severity, Peru, ENAHO 2025.
Values are survey-weighted percentages or weighted medians (IQR). Amounts are expressed as kg/person/year and are reported among households with acquisition >0. Annualized quantities were obtained from the INEI-standardized variables I601B2 and I601D2, expressed in kilograms. FIES raw-score bands indicate increasing severity and are not official FAO food-insecurity categories. FIES, Food Insecurity Experience Scale; IQR, interquartile range; ENAHO, Encuesta Nacional de Hogares (Peruvian National Household Survey).
After multivariable adjustment, increasing food-insecurity severity was primarily associated with lower amounts of fruits and vegetables acquired by households. Each 1-point increase in the FIES score was associated with 2.6% lower fruit acquisition (aMR = 0.974; 95% CI: 0.968–0.980) and 1.3% lower vegetable acquisition (aMR = 0.987; 95% CI: 0.982–0.992) among households with positive acquisition. Legume acquisition showed no significant overall association with FIES severity. Higher FIES scores were associated with lower probabilities of acquiring fish and seafood, processed meats, and carbonated soft drinks; for processed meats and carbonated soft drinks, the quantities acquired among positive-acquisition households were also modestly lower, whereas the fish and seafood quantity model was null. For cereals, pseudocereals, and cereal-based foods, the continuous models showed only minimal positive associations with acquisition probability (aPR = 1.001; 95% CI: 1.000–1.002) and amount acquired (aMR = 1.005; 95% CI: 1.001–1.009; Table 4).
Table 4
| Food group | Any acquisition, aPR (95% CI) | p | Amount acquired, aMR (95% CI) | p |
|---|---|---|---|---|
| Fruits | 1.000 (0.999–1.001) | 0.537 | 0.974 (0.968–0.980) | <0.001 |
| Vegetables | 1.002 (1.000–1.004) | 0.024 | 0.987 (0.982–0.992) | <0.001 |
| Legumes | 0.997 (0.993–1.001) | 0.163 | 0.997 (0.990–1.004) | 0.348 |
| Cereals, pseudocereals, and cereal-based foods | 1.001 (1.000–1.002) | 0.049 | 1.005 (1.001–1.009) | 0.009 |
| Fish and seafood | 0.996 (0.993–0.999) | 0.016 | 1.000 (0.992–1.007) | 0.923 |
| Processed meats | 0.972 (0.961–0.983) | <0.001 | 0.986 (0.975–0.998) | 0.022 |
| Carbonated soft drinks | 0.984 (0.977–0.992) | <0.001 | 0.988 (0.977–0.999) | 0.032 |
Adjusted associations between food insecurity severity and household food acquisition, Peru, ENAHO 2025.
aPRs correspond to the probability of any acquisition; aMRs correspond to the amount acquired among households with acquisition >0. Estimates represent the association per 1-point increase in the FIES score. Models were adjusted for sex, age, and educational attainment of the household head; household size; presence of children aged <5 years and adults aged ≥65 years; poverty status; urban/rural residence; geographic domain; and month of interview, accounting for the ENAHO complex survey design. aPR, adjusted prevalence ratio; aMR, adjusted mean ratio; CI, confidence interval; FIES, Food Insecurity Experience Scale; ENAHO, Encuesta Nacional de Hogares (Peruvian National Household Survey).
Effect-modification analyses (Panel A) showed heterogeneity for fruits, vegetables, legumes, and cereals, whereas there was no evidence that the associations for processed meats or carbonated soft drinks differed by poverty status, area of residence, or food-assistance participation (Table 1). For cereals, the small positive association with FIES was observed among non-poor households (aMR = 1.008; 95% CI: 1.004–1.012), urban households (aMR = 1.007; 95% CI: 1.003–1.011), and households not participating in food assistance (aMR = 1.011; 95% CI: 1.006–1.016), but was null or slightly inverse in the comparison strata (p for interaction = 0.010, <0.001, and < 0.001, respectively). When FIES was modeled categorically instead of continuously (Panel B), fruit acquisition showed a graded decline across increasing severity bands. Vegetable acquisition was significantly lower in the 6–8 band, while processed-meat acquisition was also significantly lower only in this highest-severity band. Carbonated soft-drink acquisition was lower in the 1–2 and 3–5 bands but was not significantly different in the 6–8 band. In contrast, cereal acquisition was modestly higher in the 1–2 and 3–5 bands but not in the 6–8 band, whereas legume acquisition remained comparatively stable (Table 1 and Figure 1). In the sensitivity analysis (Panel C), excluding observations above the food-group-specific 99.9th percentile did not materially alter the estimates.
Figure 1
4 Discussion
In this nationally representative sample of Peruvian households, food-insecurity severity was associated with a selective contraction of the household food basket rather than a uniform change across all food groups. The clearest pattern was a reduction in the quantities of fruits and vegetables acquired, while additional groups showed different responses at the probability and quantity margins. Legumes remained comparatively stable, and cereals, pseudocereals, and cereal-based foods were largely preserved rather than showing a progressive decline.
The inverse association between food insecurity severity and the quantities of fruits and vegetables acquired is consistent with previous evidence linking food insecurity to lower consumption of these foods and poorer overall diet quality (Angeles-Agdeppa et al., 2021; Hanson and Connor, 2014). In Latin America, increasing food-insecurity severity has also been associated with progressively lower adherence to dietary patterns characterized by fruits, vegetables, and animal-source protein foods (Betancourt-Núñez et al., 2023). Likewise, studies focused on food acquisition practices have documented differences in the purchase and procurement of fruits and vegetables between food-secure and food-insecure households (Ilori et al., 2024).
Our findings extend this evidence by indicating that contraction in household food acquisition may occur through the quantity margin even when a food group remains present in the household basket. This was particularly apparent for fruits, for which acquisition prevalence changed little while adjusted quantities declined across increasing FIES bands. Vegetables showed a more marked contraction at the highest severity band. Affordability is a plausible contributor because fruits and vegetables tend to have relatively high costs per unit of energy (Drewnowski, 2010), and affordability is a central dimension of access to a healthy diet (FAO, 2023). Previous ENAHO analyses have also linked household food expenditure to the nutritional quality of foods available within Peruvian households (Bonilla-Aguilar and Bernabe-Ortiz, 2023).
Greater food-insecurity severity was also associated with a lower probability of acquiring fish and seafood, processed meats, and carbonated soft drinks; processed-meat and carbonated-soft-drink quantities were also modestly lower among households that acquired them. This pattern does not support a simple or uniform substitution from nutritionally preferable foods toward energy-dense products as food insecurity worsens. In contrast to studies from high-income settings that have reported positive associations between food insecurity and consumption of sugar-sweetened beverages or processed meats (Fernández et al., 2020; Flórez et al., 2023; Leung et al., 2014), our findings in Peru suggest a different scenario. While systematic evidence on sugar-sweetened beverages remains heterogeneous (Cope and Chestnutt, 2024), the lower acquisition of these products among food-insecure households in our study aligns with previous qualitative evidence from the Peruvian Amazon, where households facing greater food-access constraints adopted complex coping strategies that included reducing some sugary products and substituting preferred fish and meats with lower-cost alternatives (Ambikapathi et al., 2018). Rather than reflecting a consistent shift toward energy-dense, nutrient-poor foods, the present findings are more consistent with a selective contraction of the household food basket, affecting several food groups, including some that are often considered "less healthy." This interpretation highlights that economic constraints may operate primarily through reduced purchasing power across multiple food categories, rather than through a specific substitution pattern toward lower-quality products. However, the cross-sectional design limits our ability to determine whether these reductions reflect deliberate household strategies, supply-side constraints, or other unmeasured factors.
Unlike fruits and vegetables, legume acquisition remained relatively stable as food-insecurity severity increased, with no significant overall association with either acquisition probability or quantity. This pattern is consistent with evidence from other low- and middle-income settings showing heterogeneous responses for this food group (Alvarez-Sanchez et al., 2021). In Peru, however, the overall stability concealed territorial heterogeneity: higher FIES severity was associated with lower legume quantities in rural households but not in urban households. Because this interaction analysis was exploratory, the finding should be interpreted as evidence of possible territorial variation rather than as a definitive subgroup effect.
Cereals, pseudocereals, and cereal-based foods showed a different pattern. Acquisition was nearly universal, and the continuous model indicated only a very small positive association between FIES severity and quantity. The categorical analysis clarified that this was not a dose-response increase: quantities were modestly higher in the intermediate FIES bands but were not different in the 6–8 band compared with households scoring 0. This pattern is more consistent with relative preservation of staple cereal-based foods than with progressive substitution toward them. Their generally lower cost per unit of energy may contribute to this preservation (Drewnowski, 2010), but the cross-sectional design does not allow displacement of other food groups by cereals to be inferred. Moreover, the cereal association varied by poverty, area of residence, and food-assistance participation, although the stratum-specific deviations from the null were small and should be interpreted as exploratory.
Effect-modification analyses further indicated that the relationship between food insecurity and acquisition was context dependent. For fruits, the inverse association was more pronounced among non-poor and urban households, whereas for vegetables it was stronger among poor and rural households. Legume heterogeneity was concentrated mainly by area of residence. In contrast, processed meats and carbonated soft drinks showed no evidence of modification by poverty, area, or food-assistance participation. For cereals, the small positive association was concentrated among non-poor, urban, and non-participating households and was absent or slightly inverse in the comparison strata. These patterns are consistent with broader evidence that socioeconomic and geographic context can shape food access (McCullough et al., 2022), but the interaction results should not be interpreted causally.
The inclusion of foods obtained through non-purchase sources is important for interpreting these results. The ENAHO acquisition measure combined purchases with own production, self-supply, in-kind payments, gifts, and social-program sources; thus, the estimated associations reflect net household acquisition across multiple pathways. In Peru, the main food assistance programs include Qali Warma (national school feeding program serving over 4.1 million children in public schools; Francke et al., 2026), Cuna Más (targeting children under 36 months in poverty, providing three daily meals; Tarazona-Meza et al., 2025), the Vaso de Leche program (providing daily food rations to vulnerable populations, including children, pregnant women, and the elderly; Rodríguez-Massa et al., 2024), and the Programa de Complementación Alimentaria (supporting community kitchens and soup kitchens across the country; MIDIS, 2026).
In Lima, food-insecure families have been shown to use both formal programs and informal resources such as food credit and food received from others, with greater use at higher levels of food insecurity (Brewer et al., 2021). Such pathways may buffer reductions in some food groups or alter the composition of foods entering the household. For instance, the relative stability of legume and cereal acquisition observed in our study could partly reflect the influence of these programs, as they often provide staple foods and fortified products. However, the present analysis did not decompose each food group by acquisition source, and food-assistance participation is targeted to vulnerable households. Consequently, the observed interaction patterns involving program participation should not be interpreted as causal effects of food assistance; program-specific evaluations would be required to establish whether and how these interventions modify household food acquisition.
4.1 Strengths and limitations
This study has several strengths. First, it used a large, nationally representative sample of Peruvian households and accounted for the complex sampling design of ENAHO, supporting the generalizability of the findings at the national level. Second, food acquisition estimates incorporated both purchased quantities and foods obtained through other sources, providing a more comprehensive approximation of foods available within the household. In addition, the two-part modeling approach distinguished between the probability of acquiring a given food group and the quantity obtained among households with positive acquisition, thereby avoiding reliance on a single dimension of food acquisition behavior. The consistency of the findings after excluding extreme values further supports their robustness.
Several limitations should also be considered. The cross-sectional design precludes establishing temporality or causal relationships between food insecurity and food acquisition. Module 601 measures household-level acquisition rather than individual dietary intake; therefore, per capita quantities standardize by household size but do not represent the distribution or actual consumption of food among household members. Accordingly, the findings should not be interpreted as direct measures of dietary diversity or overall diet quality. Although models adjusted for multiple sociodemographic, economic, geographic, and temporal characteristics, residual confounding remains possible. In addition, purchased and non-purchase quantities were combined in the primary outcomes, so the present analysis cannot determine whether observed associations were driven by market purchases, own production, transfers, or social-program sources. Similarly, because food-assistance participation is nonrandom and targeted according to vulnerability, interaction estimates involving program participation should be regarded as exploratory rather than causal.
Additionally, food acquisition quantities were self-reported and may be subject to recall bias, particularly among households with irregular acquisition patterns or those relying on multiple non-market sources. Such bias, if present, could affect the precision of the estimated quantities, although it would likely attenuate rather than inflate the associations.
Finally, prepared dishes could not be disaggregated into their individual ingredients because standardized recipes are not available in ENAHO; consequently, their components were not assigned to the food groups analyzed. This likely led to some underestimation of acquired quantities when foods from the study groups were consumed as part of mixed dishes. The magnitude of this underestimation may not have been uniform across households, given previous evidence of socioeconomic and geographic differences in the consumption of and expenditure on foods prepared outside the home, with greater participation among higher socioeconomic groups and in urban settings (Lozada-Urbano et al., 2022; Landais et al., 2022). Therefore, the omission of ingredients from prepared dishes may have resulted in greater underestimation of some food groups among urban or more affluent households. However, because the composition of these preparations is unknown, the magnitude and direction of any resulting bias in the estimated associations cannot be determined with certainty.
4.2 Public health implications
From a public health perspective, these findings highlight the need to complement traditional food insecurity indicators with measures capable of capturing changes in both the composition and quantity of household food acquisition. Monitoring only the presence or absence of specific food groups may fail to detect gradual deterioration in food access. Integrating information on food insecurity experiences with food acquisition patterns in national surveys such as ENAHO could therefore provide a more sensitive approach to monitoring food and nutritional vulnerability.
Interventions aimed at improving food security should consider not only energy sufficiency but also households’ capacity to maintain adequate quantities across key components of the household food basket. The socioeconomic and territorial heterogeneity observed in this study suggests that strategies may need to be adapted to population-specific contexts. National monitoring could also benefit from distinguishing acquisition pathways, because reductions in market purchases may be partly offset by own production, transfers, or food-assistance programs. Future longitudinal and program-specific studies are warranted to determine whether changes in household acquisition precede deterioration in diet quality and to clarify how social protection policies influence the composition and quantity of foods available to households.
5 Conclusion
Greater food-insecurity severity was associated with distinct and food-group-specific changes in household acquisition in Peru. The most consistent contraction involved lower quantities of fruits and vegetables, while acquisition of fish and seafood, processed meats, and carbonated soft drinks also declined on at least one margin of the two-part model. Legumes remained comparatively stable, whereas cereals, pseudocereals, and cereal-based foods were relatively preserved overall, without a monotonic pattern across FIES severity bands, although exploratory analyses indicated modest heterogeneity by poverty status, area of residence, and food-assistance participation. These findings are consistent with selective contraction of the household food basket rather than uniform substitution toward foods of poorer nutritional quality and highlight the importance of considering both acquisition quantity and the different monetary and nonmonetary pathways through which households obtain food.
Statements
Data availability statement
Publicly available datasets were analyzed in this study. This data can be found at: https://proyectos.inei.gob.pe/microdatos/.
Ethics statement
Ethical review and approval were not required for this study because it involved a secondary analysis of publicly available, de-identified ENAHO 2025 microdata. The original survey was conducted by Peru’s National Institute of Statistics and Informatics (INEI), which was responsible for obtaining informed consent from participants. The study was conducted in accordance with local legislation and institutional requirements.
Author contributions
JS: Writing – review & editing, Conceptualization, Methodology, Funding acquisition, Writing – original draft, Formal analysis.
Funding
The author(s) declared that financial support was received for this work and/or its publication. Universidad Peruana Unión (UPeU) provided financial support solely for the Article Processing Charge (APC) of this publication.
Acknowledgments
The author acknowledges the Instituto Nacional de Estadística e Informática (INEI) of Peru for providing public access to the ENAHO 2025 microdata used in this study.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The author JS declared that they were an editorial board member of Frontiers, at the time of submission. This had no impact on the peer review process and the final decision.
Generative AI statement
The author(s) declared that Generative AI was used in the creation of this manuscript. During the preparation of this manuscript, the author used an AI-assisted language tool to support language editing and improvement of manuscript clarity. The author subsequently reviewed and edited the content and takes full responsibility for the final version of the manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fsufs.2026.1966104/full#supplementary-material
References
1
Alvarez-SanchezC.MoltedoA.TroubatN.ManyaniT.YassinF.KeppleA.et al. (2021). The Relationship Between Food Insecurity and Dietary Outcomes. Rome, Italy: FAO.
2
AmbikapathiR.RothsteinJ. D.YoriP. P.OlorteguiM. P.LeeG.KosekM. N.et al. (2018). Food purchase patterns indicative of household food access insecurity, children’s dietary diversity and intake, and nutritional status using a newly developed and validated tool in the Peruvian Amazon. Food Secur.10, 999–1011. doi: 10.1007/S12571-018-0815-2.
3
Angeles-AgdeppaI.ToledoM. B.ZamoraJ. A. T. (2021). Moderate and severe level of food insecurity is associated with high calorie-dense food consumption of Filipino households. J. Nutr Metab., 2021. 5513409. doi: 10.1155/2021/5513409.
4
Betancourt-NúñezA.Nava-AmanteP. A.Bernal-OrozcoM. F.VizmanosB.Vargas-GarcíaE. J.Márquez-SandovalF.et al. (2023). Food insecurity was negatively associated with adherence to the “fruits, vegetables, and foods rich in animal protein” dietary pattern among university students’ households: the 2018 Mexican National Household Survey. BMC Publ. Health23, 854. doi: 10.1186/S12889-023-15755-Z.
5
Bonilla-AguilarK.Bernabe-OrtizA. (2023). Association between total available nutritional quality and food expenditure in Peruvian households, 2019–2020. Cad. Saúde Pública39:e00021923. doi: 10.1590/0102-311XEN021923, .
6
BrewerJ. D.SantosM. P.LopezM. A.Paz-SoldanV. A.ChaparroM. P. (2021). Use of formal and informal food resources by food insecure families in Lima, Peru: a mixed-methods analysis. J. Community Health46, 1069–1077. doi: 10.1007/S10900-021-00989-Y, .
7
BrunetG.FajardoG.CostaM.BonillaL.GonzálezF.BentancorS.et al. (2026). How do households cope with food insecurity? Quantitative analysis of the coping strategies of households with children attending public childcare in Uruguay. Appetite216:108267. doi: 10.1016/j.appet.2025.108267, .
8
Carvajal-AldazD.CucalonG.OrdonezC. (2022). Food insecurity as a risk factor for obesity: a review. Front. Nutr.9:1012734. doi: 10.3389/FNUT.2022.1012734
9
ColetroH. N.De Menezes-JúniorL. A. A.MendonçaR. D. D.MeirelesA. L.MacHado-CoelhoG. L. L.De MenezesM. C. (2023). The combined consumption of fresh/minimally processed food and ultra-processed food on food insecurity: COVID Inconfidentes, a population-based survey. Public Health Nutr.26, 1414–1423. doi: 10.1017/S136898002300054X, .
10
CopeA. L.ChestnuttI. G. (2024). A systematic review of the association between food insecurity and behaviours related to caries development in adults and children in high-income countries. Community Dent. Oral Epidemiol.52, 625–647. doi: 10.1111/CDOE.12959, .
11
DrewnowskiA. (2010). The cost of US foods as related to their nutritive value. Am. J. Clin. Nutr.92, 1181–1188. doi: 10.3945/ajcn.2010.29300, .
12
FAO (2016). Methods for estimating comparable prevalence rates of food insecurity experienced by adults throughout the world. Available online at: https://openknowledge.fao.org/server/api/core/bitstreams/823b2fcd-bc52-4371-a31d-ea2aa74e63d5/content (Accessed August 15, 2026).
13
FAO (2023). Global indicators on the costs of healthy diets and how many people can’t afford them FAO. Available online at: https://www.fao.org/newsroom/detail/global-indicators-on-the-costs-of-healthy-diets-and-how-many-people-can-t-afford-them/ (Accessed August 15, 2026).
14
FAO–World Bank (2018). Food Data Collection in Household Consumption and Expenditure Surveys: Guidelines for Low- and Middle-Income Countries. Available online at: https://www.worldbank.org/en/programs/lsms/publication/Food-data-collection-in-household-consumption-and-expenditure-surveys-guidelines-for-low-and-middle-income-countries? (Accessed August 15, 2026).
15
FernándezC. R.ChenL.ChengE. R.CharlesN.MeyerD.MonkC.et al. (2020). Food insecurity and sugar-sweetened beverage consumption among WIC-enrolled families in the first 1,000 days. J. Nutr. Educ. Behav.52, 796–800. doi: 10.1016/j.jneb.2020.03.006
16
FlórezK. R.AlbrechtS. S.HwangN.ChambersE.LiY.GanyF. M.et al. (2023). Household food security and consumption of sugar-sweetened beverages among New York City (NYC) children: a cross-sectional analysis of 2017 NYC kids’ data. Nutrients15:3945. doi: 10.3390/nu15183945, .
17
FranckeP.AcostaG.QuispeD. (2026). Effects of Peru’s national school feeding program (Qali Warma) on overweight and obesity among children aged 36–59 months. Obesities6: 25. doi: 10.3390/obesities6030025
18
GuptaP.NarineL. K. (2025). Coping strategies utilized by low-income households to address food security in Utah. J. Hunger. Environ. Nutr.20, 1616–1630. doi: 10.1080/19320248.2025.2547330
19
Guzman-VilcaW. C.Carrillo-LarcoR. M.Tarazona-MezaC. (2025). Consumption of fruits and vegetables among Peruvian adults: analysis of a national health survey 2017–2018. PLOS Glob. Public Health5:e0004222. doi: 10.1371/JOURNAL.PGPH.0004222, .
20
HansonK. L.ConnorL. M. (2014). Food insecurity and dietary quality in US adults and children: a systematic review. Am. J. Clin. Nutr.100, 684–692. doi: 10.3945/ajcn.114.084525, .
21
Hernández-VásquezA.Vargas-FernándezR.Guerra ValenciaJ. (2026). Food insecurity in Peru: measurement properties of the FIES and prevalence estimates from the 2025 household survey. Nutrients18: 2528. doi: 10.3390/NU18152528
22
Hernández-VásquezA.Visconti LopezF. J.Vargas-FernándezR. (2022). Socio-economic inequalities in the consumption of fruits and vegetables in Peru between 2014 and 2019. Public Health Nutr.25:3527. doi: 10.1017/S1368980022001860
23
HutchinsonJ.TarasukV. (2022). The relationship between diet quality and the severity of household food insecurity in Canada. Public Health Nutr.25, 1013–1026. doi: 10.1017/S1368980021004031, .
24
IhabA. N.RohanaA. J.Wan MananW. M.Wan SuriatiW. N.ZalilahM. S.RusliA. M. (2012). Food expenditure and diet diversity score are predictors of household food insecurity among low income households in rural district of Kelantan Malaysia. Pak. J. Nutr.11, 967–973. doi: 10.3923/pjn.2012.967.973
25
IloriT.ChristofidesN.Baldwin-RagavenL. (2024). The relationship between food insecurity, purchasing patterns and perceptions of the food environment in urban slums in Ibadan, Nigeria. BMC Nutr.10:122. doi: 10.1186/S40795-024-00929-8, .
26
INEI. (2025a). Consulta por Encuesta. Available online at: https://proyectos.inei.gob.pe/microdatos/Consulta_por_Encuesta.asp (Accessed August 15, 2026).
27
INEI. (2025b). Instituto Nacional de Estadística e Informática del Perú (INEI). Available online at: https://www.inei.gob.pe/media/MenuRecursivo/boletines/informe-de-gobernabilidad-agosto-2025.pdf (Accessed August 15, 2026).
28
KentK.SchumacherT.KocarS.SeivwrightA.VisentinD.CollinsC. E.et al. (2024). Increasing food insecurity severity is associated with lower diet quality. Public Health Nutr.27:e61. doi: 10.1017/S1368980024000417, .
29
LandaisE.Miotto-PlessisM.BeneC.Maitre d’HotelE.TruongM. T.SoméJ. W.et al. (2022). Consumption of food away from home in low- and middle-income countries: a systematic scoping review. Nutr. Rev.81:727. doi: 10.1093/NUTRIT/NUAC085
30
LeungC. W.EpelE. S.RitchieL. D.CrawfordP. B.LaraiaB. A. (2014). Food insecurity is inversely associated with diet quality of lower-income adults. J. Acad. Nutr. Diet.114, 1943–1953.e2. doi: 10.1016/j.jand.2014.06.353, .
31
Lozada-UrbanoM.HuamánF.XirinachsY.Rivera-LozadaO.Alvarez-RiscoA.YáñezJ. A. (2022). Poverty, household structure and consumption of foods away from home in Peru in 2019: a cross-sectional study. Foods11, 2547. doi: 10.3390/FOODS11172547
32
LumleyT. (2004). Analysis of complex survey samples. J. Stat. Softw.9, 1–19. doi: 10.18637/JSS.V009.I08
33
McCulloughM. L.ChantaprasopsukS.IslamiF.Rees-PuniaE.UmC. Y.WangY.et al. (2022). Association of socioeconomic and geographic factors with diet quality in US adults. JAMA Netw. Open5:E2216406. doi: 10.1001/JAMANETWORKOPEN.2022.16406, .
34
MIDIS (2026). Programa de Complementación Alimentaria (PCA) - Contenido institucional - Ministerio de Desarrollo e Inclusión Social - Plataforma del Estado Peruano. Available online at: https://www.gob.pe/11779-ministerio-de-desarrollo-e-inclusion-social-programa-de-complementacion-alimentaria-pca (Accessed August 15, 2026).
35
Rodríguez-MassaP.Lucía Cherres-BernalA.Romaní-RomaníF. (2024). Coverage of the Vaso de Leche and Qali Warma programs among children under five years of age and their associated factors. Horiz. Med. (Lima)24: 2627. doi: 10.24265/horizmed.2024.v24n4.13
36
Tarazona-MezaC.BartoliniR. M.RomeroK.PradeillesR.GoyaC.RoushamE. K.et al. (2025). Facilitators and barriers to anemia prevention in the urban government childcare program for infants and young children in Peru. Glob. Health Action18:2475580. doi: 10.1080/16549716.2025.2475580, .
Summary
Keywords
food acquisition, food basket, food insecurity, food insecurity experience scale, household survey, Peru
Citation
Saintila J (2026) Food insecurity severity, household food acquisition patterns, and food basket contraction in Peru: a national analysis of ENAHO 2025. Front. Sustain. Food Syst. 10:1966104. doi: 10.3389/fsufs.2026.1966104
Received
12 August 2026
Revised
09 September 2026
Accepted
10 September 2026
Published
24 September 2026
Volume
10 - 2026
Edited by
Hamid, Rani Lakshmi Bai Central Agricultural University, India
Reviewed by
Rônisson Thomas de Oliveira-Silva, Federal University of Pernambuco, Brazil
Abhimanyu Thakur, Dr. Yashwant Singh Parmar University of Horticulture and Forestry, India
Updates
Copyright
© 2026 Saintila.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Jacksaint Saintila, jacksaintsaintila@gmail.com
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.